Hey Reddit,
I recently saw a few article about Denoisers (-> e.g Nvidia Denoiser in OptiX, or the V-Ray Denoiser [1], [2], [3]), and I am super impressed by the technology and would love to learn more about it.
I am especially interested in where it can be applied, which devices you need for it, and its impact on the film industry. For instance, does drastically reduced render-times mean we can expect shorter time-periods between high-quality movies? 😃
Would be cool if you could help me in my search for answers. Plus, please be a little forgiving, I’m not too deep into the technical material :S (just some dude who is generally impressed by it)
Here are my thoughts:
-
Is the Denoiser ‘only’ good enough for previews (-> fast work iterations), or is it also used in the final render pass of the movie? (which output we see in the cinema)
-
If it would be good enough for the final pass, does that imply the movie industry can produce their animated films much faster now? And thus, we should expect more animated movies coming out soon? (looking at you @ Pixar 😃)
-
Do we need a GPU do use the Denoiser, or does it work with CPUs as well? I suspect it is neural-network inference, and thus should work on both devices? (if not: please provide some details)
-
Does the Denoiser work on ‘all’ render scenes that use path tracing ( ray tracing and/or radiance)? Moreover, which segments does this include? (Animation movies? Action film? GCI? Cartoons? Anime?)
-
Do all modern render-engines have their own Denoiser? If so: Which one would you say is currently the state-of-the-art?
Thanks a lot ! L0drik-
[1] https://developer.nvidia.com/optix-denoiser
[3] https://www.chaosgroup.com/blog/experiments-with-v-ray-next-using-the-nvidia-optix-denoiser
Submitted July 11, 2018 at 02:53PM by L0drik- https://ift.tt/2m7lU5T via TikTokTikk
No comments:
Post a Comment